|
(11) | EP 0 821 070 A1 |
| (12) | EUROPEAN PATENT APPLICATION |
|
|
|
|
||||||||||||||||
|
|||||||||||||||||
| (54) | Pit-1 gene polymorphism and trait selection in animals |
| (57) The present invention relates to a genetic marker used to distinguish amongst animals
a trait for milk producing capabilities or muscular beef producing capabilities said
genetic marker comprising a mutation in a fragment of a Pit-1 gene wherein after digestion
with a restriction endonuclease, three allele patterns are observed, the fully digested
pattern being indicative of a trait for muscularity in said animal, while the intermediate
digested/nondigested pattern or the nondigested pattern being indicative of a milk
producing trait in said animal. A process and kit using this genetic marker is also
disclosed. |
BACKGROUND OF THE INVENTION
Field of the invention
[0001] The present invention relates to a genetic marker associated with different conformational traits. More specifically, the present invention describes a process wherein a polymorphism in a Pit-1 gene is used to determine traits in animals such as milk production and muscularity with ease.
Description of the Prior Art
[0002] Selection of a particular trait in a mammal is presently very expensive and very slow. Usually the selection process involves a geneological evaluation of the mammals history over a long period of time. This evaluation is based on various traits of the mammal or animal such as birth weight, growth weight, build, muscle strength, firmness, marbling, color, and the like.
[0003] Most of the selection of a particular trait in an animal to date, involves visually characterizing the specific traits over a time frame or weighing the animal at particular times. The animals with the quality traits that are to be selected are then bred with similar animals such that the particular trait is hopefully dominant in the next generation or the generations to follow.
[0004] The present methods for trait selection in mammals are often tedious and open to judgment of an expert in the field, such as a breeder. However, there is never any real assurance that the choice being made will dominate over the forthcoming generations. For example, in order to select a cow that is a good milk producing animal, it takes between 36 to 48 months to make such choice and after the choice is made, it is often based on hypothesis and the breeder's judgment.
[0005] In view of the uncertainty, expense and time involved with the current methods of trait selection in animals, new methods are currently under development which methods utilize a more scientific process which will hopefully improve the selection process.
[0006] One such method is the study of candidate genes to determine whether specific genes are associated with conformational traits in mammals and therefore these genes can be used as molecular markers to select particular traits of interest. This method first requires identification of candidate genes or anonymous genetic markers associated with the traits of interest. The candidate gene approach can be successful, but first genes must be identified in the species of interest and correlated to the traits of interest.
[0007] The somatotropin system has several genes that may play a role in the control of particular traits in animals since this system is associated with growth, lactation, reproduction and immunity. The somatotropin system is quite complicated and involves at a hypothalamic level, somatocrinin and somatostatin; at a pituitary level, pituitary-specific transcription factor (Pit-1) which is responsible for growth hormone expression in mammals; at a hepatic level, growth hormone receptor and growth hormone plasmatic transport protein; and at a cellular level, growth hormone receptor, insulin-growth factor-1 and insulin growth factor transport protein.
[0008] Selection of genes from this somatotropin system is that may influence particular traits in animals is quite complicated, since this system has many different functions in different parts of the animal, from the pituitary to the cellular level.
[0009] The present invention involves the selection of a gene, the pituitary-specific transcription factor (hereinafter referred to as Pit-1) that can act as a genetic marker to characterize specific traits in animals.
[0010] Pit-1 is a member of the POU family of homeo-domain transcription factors and plays an important role in developmental processes. The POU-domain was originally identified as a highly conserved region of 150 to 160 amino acids found in three mammalian transcription factors, Pit-1, Oct-1, Oct-2 and also in the product of nematode gene unc-86 (Herr et al., Genes & Dev. 2: 1513 (1988); Ruvkun and Finnery, Cell 64:475 (1991)).
[0011] Pit-1 is a pituitary-specific transcription factor that regulates growth hormone, activates prolactin and has a role in pituitary cell differentiation and proliferation (Steinfelder et al.,P.N.A.S., USA 88:3130 (1991). Mutations in the Pit-1 gene responsible for the dwarf phenotypes of the Snell and Jackson mice and lead to anterior pituitary hypoplasia (Li et al., Nature 347:528 (1992)). Moreover, it has been shown that the inhibition of Pit-1 synthesis leads to a decrease in prolactin and growth hormone (GH) expression and to a dramatic decrease in cell proliferation in GH and prolactin producing cell lines (McCormick et al., Nature 345:829 (1990)).
[0012] In human, different mutations in the Pit-1 gene have also been reported in patients with familial pituitary hypoplasia (Pfaffle et al., Science 257:1118 (1992)); and in patients with sporadic combined pituitary hormone deficiency (Radovick et al., Science 257:1115 (1992); Tatsumi et al., Nature Genetics 1: 56 (1992).
[0013] The Association of Pit-1 polymorphisms with growth and carcass traits in pigs has been described by Yu et al., J. Anim. Sci. 73: 1282 (1995). Yu et al., supra described three Pit-1 polymorphisms in pigs based on two restriction fragment length polymorphisms (hereinafter referred to as RFLP) using a Pit-1 POU-domain cDNA probe and the restriction enzymes BamHI and MspI and a PCR/RFLP using RsaI.
[0014] Results from Lu et al.'s, supra, mixed-model analysis revealed that pigs with the MspI CC genotype were associated with heavier birth rate than the DD genotype pigs. Moreover, with the Pit-1 BamHI polymorphisms heavier birth weight was significantly associated with the BB genotype, although the authors cautioned against concluding such association since the BB genotype population was extremely small.
[0015] Although Woolard et al., J. Anim. Sci. 72:3267 1994) recognized a HinfI polymorphism at the bovine Pit-1 gene locus, these authors failed to link this mutation to the selection trait in animals. The conclusion drawn in Woolard, supra was that polymorphic fragments that were observed were consistent with autosomal Mendelian inheritance.
[0016] There is no disclosure in Yu et al. or Woolard et al of any association of the allele pattern AB with milk production, nor the allele pattern BB with muscularity in animals.
[0017] Therefore, the present invention overcomes the disadvantages of the current methods of trait selection in animals by providing a scientific basis for selection of traits by use of a genetic marker.
[0018] Moreover, the process described in the present invention can be used to characterize superior milk producing animals from animals having meat producing characteristics.
[0019] It has been surprisingly discovered that a polymorphism in the Pit-1 gene can be used to characterize traits such as milk production and muscularity in animals. Two alleles, A (not digested) and B (digested) were distinguished for the Pit-1 gene responsible for the activation of prolactin and growth hormone gene expression using a restriction site recognized by HinfI. The AA pattern was less frequent than the AB or BB pattern. The significant superiority of the Pit-1 AB pattern or AA pattern over the BB pattern was observed for milk, protein and angularity. Likewise the BB genotype pattern was associated with animal muscularity.
[0020] This discovery permits the use of the mutation in the Pit-1 gene to be utilized as a genetic marker to identify certain traits in animals. Once these p articular traits are identified, the animals can be sold at market with increased value due to their superior traits.
[0021] Accordingly, it is an object of the present invention to provide a genetic marker for trait selection in animals.
[0022] In another aspect, the present invention provides a process to characterize animals having superior milk production traits or muscularity traits.
[0023] In yet another aspect, the present invention provides genetically engineered animals that have superior milk production, angularity, fat, protein or muscularity traits. These and other objects are achieved by the present invention as evidenced by the summary of the invention, description of the preferred embodiments and the claims.
SUMMARY OF THE INVENTION
[0024] The present invention thus provides a genetic marker that can be used for trait selection in mammals.
[0025] Furthermore, the present invention provides a method to identify a polymorphism present in the Pit-1 gene which polymorphism can be utilized to select superior traits in animals for angularity, fat, muscularity, protein or milk production.
[0027] A genetic marker used to distinguish amongst animals a trait for milk producing capabilities or muscular beef producing capabilities said genetic marker comprising a mutation in a fragment of a Pit-1 gene wherein after digestion with a restriction endonuclease, three allele patterns are observed, the fully digested pattern being indicative of a trait for muscularity in said animal, while the intermediate digested/nondigested pattern or the nondigested pattern being indicative of a milk producing trait in said animal.
[0028] In a preferred embodiment of the present invention the restriction endonuclease utilized is HinfI.
[0029] In another aspect, the present invention relates to a process for detecting certain traits in an animal, said process comprising the steps of:
(1) isolating genomic DNA from an animal;
(2) isolating a fragment from said genomic DNA comprising a fragment of a Pit-1 gene;
(3) detecting a mutation in said Pit-1 gene fragment by using restriction endonucleases; and
(4) analyzing said mutation to determine a trait in said animal wherein upon analysis traits of muscularity and fat can be distinguished from milk producing traits in said animals.
[0030] In yet another aspect, the present invention relates to genetically engineered animals that have the characteristic traits described in the present invention.
BRIEF DESCRIPTION OF THE DRAWINGS
[0031] Fig. 1 is an electrophoretic gel illustrating the PCR/Restriction Fragment Length Polymorphism patterns using the restriction enzymes HinfI on the Pit-1 gene observed in Holstein-Friesian and Simmental Bulls. The sizes of digested fragments are on the left, and the patterns are at the top. Fragment length (in kilobases) was estimated relative to the DNA size markers φX174 DNA/HaeIII fragments.
DETAILED DESCRIPTION OF THE PREFERRED EMBODIMENTS OF THE INVENTION
[0032] As used herein, the "animal" encompasses all mammals, avians, and fish including but not limited to, cows, bulls, goats, pigs, sheep, chickens and the like.
[0033] The term "polymorphism" refers to the simultaneous occurrence in the population of genomes showing allelic variations as seen either in alleles producing different phenotypes or in changes in DNA affecting the restriction pattern.
[0034] As used herein the term "trait" encompasses any characteristic, especially one that distinguishes one animal from another.
[0035] The term "angularity", as used herein means an objective criteria used to identify specific traits of an animal in relation to specific measurements which can be taken on the animal's body. The measurements are taken on the animal with respect to certain morphological characteristics.
[0036] For example, to determine the angularity for a milk production trait, the pelvic bones and muscles surrounding the pelvic bones of an animal are measured to determine whether they are projecting or not. A scale can then be established. When the bones are very projecting, there are very little rounded muscles and thus the animals are milk producing. To the contrary, when the bones are not protruding and there are a lot of rounded muscle present on the animal, the animal would not be considered a good milk produce, but rather a beef producer.
[0037] As used herein, the term "muscularity" encompasses animals that are better beef producers that can be slaughtered for their meat than milk producers.
[0038] More specifically, the present invention relates to the use of a Pit-1 gene polymorphism as a potential marker for genetic variations in animals. Pit-1 codes for a factor of transcription in a cell and any mutation of this gene can alter by diminution or augmentation the capacity of transcription thus resulting in polymorphisms which effect the outcome of different traits in an animal.
[0039] The Pit-1 gene was previously identified in a 13-kb bovine genomic library by Woolard et al., supra. A 13-kb clone was isolated from this library by using a bovine Pit-1 cDNA, which is labeled, as a probe of:
5'-AAACCATCATCTCCCTTCTT-3' (SEQ. ID. NO: 1)
5'-AATGTACAATGTGCCTTCTGAG-3' (SEQ. ID. NO:2).
Characterization of XhoI, HinfI and EcoRI subclones of this 13-kb insert by restriction enzyme digestion and sequencing identified this clone as a bovine Pit-1 genomic fragment.[0040] Similarly methods as taught by Woolard et al, supra can be used to identify the Pit-1 gene in different genomic libraries other than bovine. This will permit the identification of specific sequences within the Pit-1 genomic fragment that can be used to amplify this sequence from different animals as described below.
[0041] The first step in identifying a mutation in the Pit-1 gene in an animal is to obtain a sample from the animal such as, but not limited to semen, blood, cells, biopsy tissues, feces and the like. Genomic DNA can then be extracted for the specimens obtained using methods known in the art as described by Sambrook et al., Molecular Cloning, A Laboratory Manual, second edition 1989.
[0042] However, it is preferable to extract the genomic DNA using the procedure described in Walsh, Biotechniques, 10:506 (1991) for semen or the procedure for blood as described by Lewin and Stewart-Haynes Biotechniques, 13:522.
[0043] After extracting the genomic DNA there are several known methods in the art to detect the mutation in the Pit-1 fragment. Any detection method can be utilized to detect the mutation. Examples of these methods include, but are not limited to RFLP, SSCP, DGGE, CFLP and single base mutations as described by Prosser, Trends Biotech 11:238-246 (1993) and Sambrook et al.,supra. These methods will be discussed in greater detail below.
[0044] For example, in the RFLP (restriction fragment length polymorphism) method, PCR primers are used to amplify by standard procedures a fragment that includes the Pit-1 gene. Any PCR primers can be utilized that would permit the amplification of the Pit-1 sequence and the method in isolating the particular clone which would identify such primers.
[0045] In a preferred embodiment of the invention, the PCR primers can be designed from intron V and exon 6 of a fragment containing the polymorphism of the Pit-1 gene, such as the 451-bp fragment described by Woolard et al.,supra. In a more preferred embodiment of the present invention, the PCR primers are as follows:
5'-AAACCATCATCTCCCTTCTT-3' (SEQ. ID. NO: 1)
5'-AATGTACAATGTGCCTTCTGAG-3' (SEQ. ID. NO:2)
[0046] Amplification of the Pit-1 fragment can be performed using standard PCR procedures, as described in Sambrook et al.,supra. It is preferable, however, to amplify the genomic DNA in a 50µl reaction volumes containing 2 mM MgCl2.
[0047] In a preferred embodiment of the invention, the following conditions for the PCR reaction can be employed: between 88°C to 98°C for 10 to 15 minutes; and between 90°C to 100°C for about 1 minute, followed by between 25 to 50 cycles at between 90°C to 100°C for 20 to 40 seconds; 40°C to 60°C for 1 to 5 minutes; and 68°C to 80°C for about 1 to 5 minutes. The last step may encompass a cycle at between 68°C to 80°C for 8 to 12 minutes.
[0048] After amplification the particular mutation in Pit-1 is then cut using various restriction enzymes or endonucleases known in the art. These restriction enzymes include, but are not limited to BamHI, EcoRI, SmaI, HinfI and the like. See, for example those enzymes described in Sambrook et al.,supra. In a preferred embodiment of the present invention with the respect to the identification of milk production, fat, protein and muscularity traits in animals, HinfI is utilized.
[0049] After digestion, the sample is then electrophoresed on agarose gels and identified with a stain such as, for example ethidium bromide, however any stain can be used that identifies the fragments
[0050] SCCP(single stranded conformation polymorphism) is also a method known in the art that can identify a mutation or mutations in the isolated genomic Pit-1 fragment. This method is based on PCR amplification, using similar primers as those described above. The amplified fragment is then labeled with a label such as 32P or with any other appropriate radioactive label. The radiolabeled fragment is then denatured, for example by heating and then subjected to quick cooling. After cooling, the fragment is then electrophoresed using non-denatured technique and then audioradiographed.
[0051] DGGE (denaturing gradient gel electrophoresis) is yet another method to detect the Pit-1 mutation. In this process, the fragment is amplified by PCR using appropriate primers, such as those described above and subjected to a denaturing gradient. The sample is further electrophoresed and the mutation is detected.
[0052] Yet another method that can be used to detect the mutation is CFLP (cleavage fragment length polymorphism). This method can detect mutations of a sole base in the DNA sequence between two molecules of wild-type DNA and of a mutant type of DNA. This method is now marketed by Boehringer Mannheim and can be purchased in the form of a kit.
[0053] Although many detection methods for mutations are available, the present invention is not limited to the methods discussed above and encompasses all methods for detecting a mutation.
[0054] The alleles and allelic patterns are then identified and statistical analysis is then performed to determine the specific traits evidenced by the identification of the alleles. More specifically, any statistical program that can identify daughter yield variations (DYD) and deregressed proofs (DRP) can be utilized. It is preferable to perform the statistical analysis using the MIXED procedure of SAS ( User's Guide:Statistics, Version 6, 4th ed. SAS Inst., Inc. Cary, N.C. (1990); Technical Report P 229 SAS Inst., Inc., Cary, N.C. (1992). The statistical analysis used in the present invention is discussed in detail in the examples below.
[0055] Also encompassed by the present invention is a kit containing extraction materials for genomic DNA, the PCR primers having SEQ ID NOS. 1 and 2 (illustrated above), the materials necessary to visualize the mutation such as electrophoretic gels and the like. The content of the kit may vary depending upon the detection methods utilized, which are discussed in detail above.
[0056] In order to further illustrate the present invention and advantages thereof, the following specific examples are given, it being understood that the same are intended only as illustrative and in nowise limitative.
EXAMPLES
1. DNA EXTRACTION AND PCR
[0057] Genomic DNA of 89 commercially available registered Italian Holstein-Friesian bulls was extracted from semen as described by Lucy et al., Domest. Anim. Endocrinol. 10:325 (1993).
[0058] The RFLP at the Pit-1 gene using HinfI restriction enzyme was revealed by PCR analysis adapted from Woolard et al., supra.
[0059] The PCR primers were designed from intron V and exon 6. The sequences of the primers used were 5'-AAACCATCATCTCCCTTCTT-3' (SEQ ID NO:1) and 5'-AATGTACAATGTGCCTTCTGAG-3' (SEQ ID. NO:2). These primers were used to amplify by standard procedures a 451-bp fragment form the genomic DNA in 50-µL reaction volumes containing 2 mM MgCl2. Conditions were 94.5°C, 10 min., and 94°C, 1 min., followed by 35 cycles of 95°C, 30 s, 56°C, 1 min., and 72°C, 2 min. The last step was 72°C for 10 min. PCR products were digested with HinfI and electrophoresed on 2% agarose gels with 1 µg/mL ethidium bromide (Figure 1).
[0060] Daughter yield deviations (DYD) computed in March 1996, were obtained from the Holstein-Friesian bulls from the Italian Holstein-Friesian Breeder Association ANAFI (Associazione Nazionale Allevatori Frison Italiana, Cremona, Italy). DYD values are not computed for fat and protein percentage as those traits are only evaluated indirectly out of solutions for yield traits and mean population values for those traits. Therefore DYD values were computed using the same approach as for the computation of genetic values for percentage traits.
[0061] Similar DYD were also not available for type traits, therefore genetic values were transformed to deregressed proofs (DRP) (Banos et al., Interbull Annual Meeting, Aarhus, Denmark, Bulletin No. 8, 1993, Sigbjorn et al., J Dairy Sci, 78:2047 (1995) that can then be considered approximate DYD.
[0062] Means and standard deviations of DYD for milk production traits and or DRP for conformation traits of the bulls sample are presented in Table I. Effective number of daughters, which is a measure of the number of daughters adjusted for their distribution inside herds was available for yield traits, but not for type traits. It was therefore approximated using the following formula:
.
TABLE 1
| Mean daughter yield deviations for milk traits and deregressed proofs for conformation traits of 89 Holstein-Friesian bulls. | ||||
| Trait | SD | Minimum | Maximum | |
| Milk traits | ||||
| Milk, kg | +317 | 221 | -231 | +899 |
| Fat, kg | +10.8 | 8.2 | -14 | +28 |
| Protein, kg | +11.6 | 7.2 | -7 | +32 |
| Fat, %1 | -0.003 | 0.091 | -0.17 | +0.23 |
| Protein, %1 | +0.021 | 0.045 | -0.11 | +0.12 |
| Effective daughters2 | 490 | 1443 | 69 | 10298 |
| Conformation traits3 | ||||
| Final score | +0.147 | 0.438 | -0.75 | +1.19 |
| Stature | +0.210 | 1.536 | -3.76 | +4.56 |
| Strength | +0.218 | 1.662 | -3.68 | +3.46 |
| Body depth | +0.340 | 1.599 | -3.42 | +3.64 |
| Angularity | +0.681 | 1.215 | -3.44 | +3.42 |
| Rump angle | -0.111 | 1.807 | -4.44 | +4.10 |
| Rump width | +0.007 | 1.591 | -3.28 | +4.34 |
| Rear legs | +0.203 | 2.266 | -5.66 | +5.66 |
| Feet | +0.053 | 1.746 | -5.26 | +3.84 |
| Fore udder | +0.038 | 2.207 | -5.44 | +5.46 |
| Heigth rear udder | +0.458 | 1.856 | -3.64 | +4.44 |
| Width rear udder | +0.864 | 1.474 | -2.80 | +4.18 |
| Udder support | +0.514 | 2.453 | -10.72 | +7.12 |
| Udder depth | -0.282 | 1.702 | -5.78 | +3.74 |
| Teat placement | +0.479 | 1.633 | -4.12 | +3.82 |
| Teat length | +0.416 | 2.112 | -4.60 | +6.74 |
| Effective daughters4 | 195 | 471 | 18 | 3199 |
| 1 Percentage fat and protein daughter yield deviations computed from yields. | ||||
| 2 Number of effective daughters for yield reported by ANAFI. | ||||
| 3 Deregressed proofs for final score reported on original scale, for linear scores on relative scale. | ||||
| 4 Approximate number of effective daughters obtained from numbers of daughters and herds. |
Statistical analysis
[0063] Statistical analysis was performed using the MIXED procedure of SAS supra. The mixed model used was
Where
[0064] y = vector of DYD or DRP of bulls; b = vector of fixed effects associated with Pit-1 pattern, u = vector of random additive polygenic effect of bulls, and e = vector of random residual effects. This model was solved using the following mixed model equations:
where A is the additive relationship matrix between the 89 bulls constructed using all known relationships (1842 known ancestors),
where D is assumed to be a diagonal matrix with the number of effective daughters for every bull on its diagonal. This matrix is then divided by the estimate of the residual variance ô2 e. This is a REML estimated (Patterson and Thompson,Biometrika 58:545 91971), here identical to non-interactive minimum variance quadratic unbiased estimation (Rao, J. Mult. Anal. 1:445 (1971), as convergence occurs after 1 round. The estimate found has the property of being the quadratic forms minimizing the sampling variance. Two assumption were made, no residual covariances between DYD or DRP and heritabilities (h2) of DYD or DRP equal to heratibilities use for genetic evaluations with the exception of percentage of fat and protein where 0.50 was assumed to be the heritability (Table 2). This method tends to overestimate additive heritability as variance due to sires is not reduce for the presence of the Pit-1 pattern in the model, but this overestimation should be not very important.
TABLE 2
| Assumed heritabilities and milk traits and conformation traits of Italian Holsteins. | |
| Trait | Heritability |
| Milk traits | |
| Milk, kg | 0.25 |
| Fat, kg | 0.25 |
| Protein, kg | 0.25 |
| Fat,%1 | 0.50 |
| Protein, %1 | 0.50 |
| Conformation traits | |
| Final score | 0.15 |
| Stature | 0.38 |
| Strength | 0.29 |
| Body depth | 0.31 |
| Angularity | 0.31 |
| Rump angle | 0.25 |
| Rump width | 0.29 |
| Rear legs | 0.16 |
| Feet | 0.18 |
| Fore udder | 0.15 |
| Heigth rear udder | 0.20 |
| Width rear udder | 0.24 |
| Udder support | 0.15 |
| Udder depth | 0.29 |
| Teat placement | 0.22 |
| Teat length | 0.22 |
| 1 Percentage fat and protein daughter yield deviations computed from yields, therefore assumed heritability is not the heritability used for breeding value estimation. |
[0065] Linear contrasts were constructed as differences between pattern solutions. Testing of contrasts was done using the following statistic:
where I'b represents differences between pattern solutions, I being the linear contrast vector, Cbb an estimate of the block of the generalized inverse of the coefficient matrix associated with pattern effects and (I'CbbI)-1 is the inverse of the squared standard error of the linear contrast. the numerator degree of freedom was approximated using rank(I) = 1. The denominator was put to n - rank(X) = 86 where n is the number of observations.
[0066] It is not certain that the presence of a given pattern has only one major effect. Therefore the following strategy based of Weller et al., J. Dairy Sci. 73: 2525 (1990) was used to test this hypothesis.
1. Traits showing single-trait significant contrasts between patterns were grouped, eventual related traits were also included.
2. Weighted correlation V and covariance P matrixs among these traits were obtained.
3. A canonical transformation was defined as
, where E is a diagonal matrix of eigenvalues, and Q a matrix of eigenvectors.
4. The transformation matrix T was defined as Q-1S where S is a diagonal matrix of the inverse standard-deviations of the original
traits, therefore
.
5. The transformation matrix was used to transform the related traits to unrelated canonical traits.
6. Approximate heritabilities and weights for the canonical traits were obtained as weighted averages of the values for the initial traits, weighting coefficients were the squared values of Q-1.
7. Canonical traits were analyzed using the methods described above for initial traits. Canonical traits showing only low relative eigenvalues explain little of the observed variance.
8. Multiple-trait linear contrasts for original effects can be estimated using back transformation of significant canonical contrasts.
9. The results for these new traits are then useful to determine if only one effect of the Pit-1 pattern can be observed, or if there are more than one significant effects. Backtransformed contrasts reflect the significant differences between original traits based on a given effect of Pit-1 on the canonical trait.
RESULTS
PCR/RFLP
[0067] The PCR product was 451 bp in length. Digestion of the PCR product with HinfI revealed two alleles: the A allele not digested with HinfI and yielding a 451 bp fragment and the B allele cut at one restriction site and generating two fragments of 244 and 207 bp in length as described by Woollard et al.,supra (Figure 1).
Relationship of PCR/RFLP to Milk production
[0068] The frequencies of the three pattern AA, AB, and BB were 2.2%, 31.5% and 66.3%. The frequencies of the A and B alleles were estimated by a maximum likelihood approach with 18.8% for A and 81.2% for B.
[0069] Table 3 shows the linear contrasts and standard errors between the three Pit-1 pattern. Therefore the highly significant contrasts (P < 0.01) observed for rear legs seem to be more due to the fact that the typed AA animals are extreme on this trait than to a real biological reason. Highly significant contrasts between AB and BB patterns were found for milk and protein yield (P < 0.01). Significant contrasts were observed for fat percentage and angularity ( P < 0.05). The AB pattern or AA pattern was superior for milk, protein yield and angularity and inferior for fat percentage. These results can be interpreted as resulting from a single positive action of the heterozygote AB or AA on milk yield, thereby influencing protein yield positively and not fat yield which gives the observed negative influence on fat percentage. The influence of Pit-1 on angularity is in this context not very surprising as this linear trait is considered being strongly related to milk yield.
TABLE 3
| Linear contrasts (C) and standard errors (SE) between the three Pit-1 patterns observed on 89 Holstein-Friesian bulls. | ||||||
| Trait | Contrast | |||||
| AA-AB1 | AA-BB1 | AB-BB | ||||
| C | SE | C | SE | C | SE | |
| Milk traits | ||||||
| Milk, kg | -152 | 156 | -21 | 150 | 131 ** | 49 |
| Fat, kg | 5.0 | 5.7 | 5.4 | 5.7 | 0.4 | 1.8 |
| Protein, kg | -4.2 | 4.9 | 0.8 | 4.5 | 4.9 ** | 1.5 |
| Fat, %2 | 0.114 | 0.062 | 0.067 | 0.062 | -0.047 * | 0.019 |
| Protein, %2 | 0.005 | 0.034 | 0.015 | 0.031 | 0.010 | 0.010 |
| Conformation traits3 | ||||||
| Final score | -0.376 | 0.299 | -0.253 | 0.298 | 0.123 | 0.092 |
| Stature | -0.745 | 1.043 | -0.501 | 1.044 | 0.244 | 0.329 |
| Strength | 0.915 | 1.143 | 1.012 | 1.138 | 0.097 | 0.367 |
| Body depth | 0.108 | 1.076 | 0.562 | 1.061 | 0.454 | 0.332 |
| Angularity | -0.478 | 0.809 | 0.072 | 0.716 | 0.550 | 0.252 * |
| Rump angle | -0.211 | 1.219 | -0.514 | 1.286 | -0.303 | 0.398 |
| Rump width | 0.019 | 0.608 | 0.147 | 1.039 | 0.128 | 0.330 |
| Rear legs | -4.404 **1 | 1.548 | -4.784 **1 | 1.542 | -0.380 | 0.479 |
| Feet | 1.588 | 1.264 | 1.731 | 1.259 | 0.142 | 0.395 |
| Fore udder | -0.653 | 1.540 | -1.256 | 1.546 | -0.603 | 0.478 |
| Heigth rear udder | -0.974 | 1.290 | -0.998 | 1.288 | -0.024 | 0.750 |
| Width rear udder | -0.378 | 1.047 | 0.072 | 2.273 | 0.449 | 0.324 |
| Udder support | -1.798 | 1.707 | -1.157 | 1.706 | 0.641 | 0.525 |
| Udder depth | -1.447 | 1.245 | -1.673 | 1.240 | -0.226 | 0.388 |
| Teat placement | -1.385 | 1.158 | -1.548 | 1.154 | -0.163 | 0.356 |
| Teat length | 0.041 | 1.297 | 0.312 | 1.396 | 0.271 | 0.446 |
| 1 Only 2.2 % of the animal were AA, therefore all results comparing this pattern are preliminary | ||||||
| 2 Percentage fat and protein daughter yield deviations computed from yields. | ||||||
| 3 Deregressed proofs for final score reported on original scale, for linear scores on relative scale. | ||||||
| * P < 0.05 | ||||||
| ** P < 0.01 |
[0070] In order to test the hypothesis of a single action we performed a canonical transformation of milk fat and protein yields. Yields were analyzed as percentage DYD were obtained as functions of yields; therefore this results in no new information. Angularity was added. The phenotypic correlation matrix was computed. Observations were weighted using the number of effective daughters. Since these numbers were different for yield and type traits approximate weights were obtained as weighted means of numbers of effective daughters. Table 4 gives the correlations. Correlations among yield traits showed the expected values with higher correlations between milk and protein than between fat and one of the other traits. Angularity showed correlations between 0.42 and 0.51 with yields traits.
TABLE 4
| Correlations among daughter yield deviations for the milk traits and angularity. | ||||
| Trait | Trait | |||
| Milk yield | Fat yield | Protein yield | Angularity | |
| Milk yield | 1.00 | 0.72 | 0.90 | 0.42 |
| Fat yield | 1.00 | 0.76 | 0.51 | |
| Protein yield | 1.00 | 0.48 | ||
| Angularity | 1.00 | |||
[0071] Results from the canonical decomposition of the correlation matrix are in table V. The first and the second canonical trait explain 90% of the total variance. Especially the last canonical trait was not very informative. Table 5 gives also the eigenvectors and the relative importance of the different traits in each eigenvector. The first canonical trait is a combination of all four traits with relative influences between 15% for angularity and 30% for protein. The second canonical trait however is more specifically linked to angularity with a relative importance of 81% in this trait. The third is associated with fat and less with milk, the fourth only with milk and protein.
TABLE 5
| Standardized eigenvectors and eigenvalues of the four canonical traits (between bracketts relative importance of eigenvalues in total variance and of values in eigenvectors in canonical traits). | |||||
| Canonical Trait | Eigenvalue | Eigenvector | |||
| Milk yield | Fat yield | Protein yield | Angularity | ||
| 1 | 2.94 (73%) | 0.532 (28%) | 0.515 (27%) | 0.548 (30%) | 0.389 (15%) |
| 2 | 0.67 (17%) | 0.349 (12%) | 0.047 (<1%) | 0.257 (7%) | -0.900 (81%) |
| 3 | 0.30 (8%) | 0.396 (16%) | -0.853 73%) | 0.283 (8%) | 0.189 (4%) |
| 4 | 0.09 (2%) | 0.662 (44%) | 0.072 (<1%) | -0.744 (55%) | 0.048(<1%) |
[0072] Table 6 shows the linear contrasts and standard-errors observed for the four canonical traits. Against the expectations the first and the second canonical traits were found very highly significant (P < 0.001) and the fourth was slightly significant (P < 0.05) for the contrasts between the AB and BB pattern. This result showed that Pit-1 could have more than one action. The first canonical trait is more specifically linked to angularity. The last trait reflected the equilibrium between milk and protein yields. In order to make these contrasts more understandable, table 7 gives the values of the contrasts and the standard errors expressed on the original scales. We observed that the backtransformed contrasts were very important for milk, fat and protein for the first canonical contrast. All were also positive with AB animals superior to BB animals. For the second canonical trait the AB were inferior for milk, fat and protein and superior for angularity. This indicates again that the influence of Pit-1 on angularity seems to be important, first through the link between yields and angularity, but also directly on angularity with a slightly negative influence on yields. Canonical trait three did not show significant contrasts and canonical trait four, despite being significant, explained only very little of the total variance. After grouping all the significant canonical traits together, we observed higher grouped contrasts as in the single-trait situation. This was especially clear for fat yield and angularity, but also for milk and protein. The reason seems to be that the multiple-trait contrasts include information from the correlated traits, especially for fat and angularity this could explain the differences. Standard errors of contrasts did not increase in an important way, they were even reduced for milk and fat yields.
TABLE 6
| Linear contrasts (C) and standard errors (SE) between the three Pit-1 patterns for the four canonical traits observed on 89 Holstein-Friesian bulls. | ||||||
| Canonical trait | Contrast | |||||
| AA-AB1 | AA-BB1 | AB-BB | ||||
| C | SE | C | SE | C | SE | |
| 1 | -0.093 | 0.098 | 0.023 | 0.093 | 0.116*** | 0.017 |
| 2 | 0.003 | 0.102 | -0.032 | 0.052 | -0.035*** | 0.009 |
| 3 | -0.021 | 0.038 | -0.016 | 0.037 | 0.005 | 0.007 |
| 42 | 0.005 | 0.021 | -0.004 | 0.021 | -0.009* | 0.004 |
| 1 Only 2.2% of the animal were AA, therefore all results comparing this pattern are preliminary. | ||||||
| 2 Eigenvalue associated with canonical trait 4 was very low, therefore the results should be interpreted as non-significant. | ||||||
| * P < 0.05 | ||||||
| *** P < 0.001 |
[0073]
TABLE 7
| Linear contrast (C) and standard error of contrast (SE) between AB and BB obtained by backtransformation on 89 Holstein-Friesian bulls. | ||||||||||
| Trait | Canonical trait | |||||||||
| 1*** | 2*** | 3 | 4* | All significant1 | ||||||
| C | SE | C | SE | C | SE | C | SE | C | SE | |
| Milk yield | 289 | 44 | -57 | 15 | 9 | 13 | -27 | 11 | 205 | 48 |
| Fat yield | 12.6 | 1.9 | -0.3 | 0.1 | -0.9 | 1.2 | -0.1 | 0.1 | 12.1 | 1.9 |
| Protein yield | 8.6 | 1.3 | -1.2 | 0.3 | 0.2 | 0.3 | 0.9 | 0.4 | 8.3 | 1.4 |
| Angularity | 1.126 | 0.169 | 0.782 | 0.211 | 0.223 | 0.032 | -0.010 | 0.004 | 1.897 | 0.27 |
| 1 | ||||||||||
| 1 Combined linear contrast using the three significant canonical traits. | ||||||||||
| * P < 0.05 | ||||||||||
| *** P < 0.001 |
CONCLUSIONS
[0074] Two alleles were distinguished for the Pit-1 gene, the growth hormone factor-1/pituitary-specific transcription factor responsible for the activation of prolactin and GH gene expression, using a restriction site recognized by HinfI. Two allele were observed, A not digested and B showing this site. The AA pattern was less frequent than the AB or BB pattern. The significant superiority of the Pit-1 AB pattern or the AA pattern over BB was observed for milk, protein and angularity. This indicates that the heterozygote animals have higher productions and greater dairyness. The fat percentage was found to be lower for AB than for BB animals, a result that results from higher milk by near constant fat yield.
[0075] These results show a single action of Pit-1. But, by using a canonical transformation approach it was observed that at least two different actions of Pit-1; one on yields and angularity and another only on angularity. These results can be explain that Pit-1 has more than one role through the activation of prolactin and the GH gene expression. A first role is influencing milk, protein (and fat) yields, a second role is linked to the muscular development of the animals, meaning the presence of AB reducing the muscularity through an improvement of angularity.
[0076] Interesting enough, these findings show the usefulness of the canonical transformation to distinguish between effects on related traits. The association of Pit-1 polymorphism and milk traits in dairy cattle was shown on the original, but also on a transformed scale. Relationships were less important for conformation traits, except angularity, a trait that is related to milk yield. Again canonical transformation showed that effects on angularity were only partially a direct consequence of influence of Pit-1 on milk traits.
1. A genetic marker used to distinguish amongst animals a trait for milk producing capabilities
or muscular beef producing capabilities said genetic marker comprising a mutation
in a fragment of a Pit-1 gene wherein after digestion with a restriction endonuclease
and visualization, three patterns of alleles are observed in which one of said allele
patterns was fully digested and being indicative of a trait of muscularity in said
animal, while the two other allelic patterns, one being digested and non/digested,
while the other being nondigested/nondigested being indicative of a milk producing
trait in said animal.
2. The genetic marker according to claim 1, wherein the restriction endonuclease used
is Hinf1.
3. The genetic marker according to Claim 1 or Claim 2 wherein, the visualization is via
an electrophoretic gel.
4. The genetic marker according to any one of claims 1 to 3 ,wherein the muscularity
trait has the BB allelic frequency.
5. The genetic marker according to any one of Claims 1 to 3, wherein the milk-producing
trait has the AB allelic frequency or the AA allelic frequency.
6. A method for detecting certain traits in an animal, said method comprising the steps
of:
(1) isolating genomic DNA from an animal;
(2) isolating a fragment from said genomic DNA comprising a fragment of the Pit-1 gene;
(3) detecting a mutation in said Pit-1 gene fragment using a restriction enzyme; and
(4) analyzing said mutation to determine traits in said animal wherein upon analysis traits of muscularity and fat are distinguished from milk producing traits in said animal.
7. The method according to Claim 6, wherein said restriction enzymes is HinfI.
8. The method according to Claim 6 or Claim 7, wherein said fragment of the Pit-1 gene
is isolated using PCR primers.
9. The method according to Claim 8, wherein said PCR primers are:
5'-AAACCATCATCTCCCTTCTT-3' (SEQ ID NO:1)
5'AATGTACAATGTGCCTTCTGAG-3' (SEQ ID NO 2).
10. The method according to any one of Claims 1 to 9, wherein the detection step further
comprises after digestion, analysis via RFLP or CFLP or SSCP or DGGE.